Paul<p><strong>Struggling to install Whisper models for Kdenlive’s smart transcription plugin? Try this…</strong></p><p>First, credit where credit is due: the following solution was cobbled together combining both advice from Kdenlive developers and a tutorial by <a href="https://vkc.sh/" rel="nofollow noopener noreferrer" target="_blank">Veronica Explains</a> linked below.</p><p>So, if you are having a hard time installing <em>Whisper</em> models on Kdenlive’s latest Flatpak package, here’s how I did it.</p><p>Stop Kdenlive and open a terminal window.</p><p>From the command line, run:</p><pre><code>flatpak run --command=/bin/bash org.kde.kdenlive</code></pre><p>That 👆 was blatantly copied from the <a href="https://vkc.sh/speech-recognition-and-automatic-captioning-on-the-kdenlive-flatpak/" rel="nofollow noopener noreferrer" target="_blank">Veronica Explains</a>‘ blogpost mentioned above that covers something related.</p><p>To quote her:</p><blockquote><p>To break that down: <code>flatpak run</code> invokes the flatpak program to run an application. You can use that to run your Flatpak programs directly from the terminal (useful when running a window manager or building a startup script).</p><p>The program we’re running is <code>org.kde.kdenlive</code>, which is the application ID for the Kdenlive program.</p><p>In between <code>flatpak run</code> and <code>org.kde.kdenlive</code>, we have <code>--command=/bin/bash</code>, which will tell Flatpak that we want to run the bash prompt inside the Kdenlive Flatpak runtime, the sandboxed environment available to the Flatpak version of the Kdenlive application.</p></blockquote><p>Hats off to Veronica for explaining things so well.</p><p>Anyway… You won’t see any difference when you jump into the sandboxed environment—no warning will be displayed and the prompt will remain unchanged.</p><p>Trust me, a no message is a good message: everything will be working as it should.</p><p>From inside the environment run Whisper on an audio or video file, any audio or video file:</p><pre><code>$HOME/.var/app/org.kde.kdenlive/data/kdenlive/venv/bin/whisper [_some video or audio file_]</code></pre><p>This will make Whisper automatically download the default model (which is <code>turbo</code>) and install it, before proceeding to transcribe the spoken bits of the file’s audio track.</p><p>Note that you can install other models with a slight variation:</p><pre><code>$HOME/.var/app/org.kde.kdenlive/data/kdenlive/venv/bin/whisper --model large [_some video or audio file_]</code></pre><p>Apart from <code>turbo</code> and <code>large</code>, you have a choice between <code>tiny</code>, <code>tiny.en</code>, <code>base</code>, <code>base.en</code>, <code>small</code>, <code>small.en</code>, <code>medium</code> and <code>medium.en</code>. That said, <code>turbo</code> should serve you just fine for most cases.</p><p>Once downloaded and transcribed, Whisper generates between one and several text/subtitle files in the current directory. If You don’t need them, you can safely remove them.</p><p>You can stop the virtual environment by typing</p><pre><code>exit</code></pre><p>And then start Kdenlive normally.</p><p>Check the models in <em>Settings</em> > <em>Configure Kdenlive…</em> > <em>Plugins</em> > <em>Speech to text</em> and clicking on the <em>Model</em> drop down. Tada!</p><p>Interestingly, if you now try downloading more models from inside Kdenlive, it will work flawlessly.</p><p>I guess it just needs that little nudge.</p><p><a rel="nofollow noopener noreferrer" class="hashtag u-tag u-category" href="https://quickfix.es/tag/captions/" target="_blank">#captions</a> <a rel="nofollow noopener noreferrer" class="hashtag u-tag u-category" href="https://quickfix.es/tag/kde/" target="_blank">#KDE</a> <a rel="nofollow noopener noreferrer" class="hashtag u-tag u-category" href="https://quickfix.es/tag/kdenlive/" target="_blank">#kdenlive</a> <a rel="nofollow noopener noreferrer" class="hashtag u-tag u-category" href="https://quickfix.es/tag/linux/" target="_blank">#linux</a> <a rel="nofollow noopener noreferrer" class="hashtag u-tag u-category" href="https://quickfix.es/tag/open-source/" target="_blank">#openSource</a> <a rel="nofollow noopener noreferrer" class="hashtag u-tag u-category" href="https://quickfix.es/tag/software/" target="_blank">#software</a> <a rel="nofollow noopener noreferrer" class="hashtag u-tag u-category" href="https://quickfix.es/tag/technology/" target="_blank">#technology</a> <a rel="nofollow noopener noreferrer" class="hashtag u-tag u-category" href="https://quickfix.es/tag/videoediting/" target="_blank">#videoediting</a></p>